|
|
|
|
|
directness and an involvement that no other amp that I have heard, tube or transistor, comes close to. And of course Ralph is a true gentleman and a constant pleasure to deal with.
You made a remark somewhere about the 6BX7 being more linear than the 6BL7 because the grid is further from the cathode, to give the lower mu. I am certainly convinced that, all other things being equal, greater grid-cathode distance is goodness, because reducing grid-cathode distance is the principal cause of "tuck under" in the plate curves at increasing Vg. However mu is determined by grid pitch and by plate-grid spacing. Even in the more sophisticated formulae developed in the 40s and 50s g-cathode distance does not have significant influence. Incidentally using a 6BX7 as the output driver of the M60 instead of the standard 6SN7, running at a higher standing current, gives a noticeable improvement in transparency and clarity.
Evidently like yourself, I have spent a lot of time looking at the physics of plate curves, Originally my idea was to come up with more accurate and physically-based Spice models, although I have become a bit distracted from that. The whole business of tuck-under, and the closely related question of variation in mu value, is not dealt with at all in any of the classic tube texts (Dow, Spangenburg, etc). Even in the journals I have found only one attempt to deal with it (The Calculation of Amplifier Valve Characteristics, Liebmann, JIEE Vol 93 1946 Part III, p138). It is clear however that the primary cause is the non-uniform field at the cathode as the grid becomes more negative. There are some other deviations from the theory which remain difficult to understand. I am intrigued by your accurate tube modeling software although I imagine you are not about to disclose too much about how it is done.
Again, thanks and congratulations for a very nice piece of work, and long may it continue.
John H
|
|
|
|
|
|
|
|
|
|
Thank you for a great letter. Nothing could be better for the tube renaissance than to have more thought put into how a tube actually works, and then to continue with the evolution of the vacuum tube. Although I welcome the reproduction of the old greats, I would like to see a new great tube. (Yes, this does mean that I do not believe that the 300B is beginning and end of tube design.) Maybe a good starting point would be a new triode based on the 8417 cathode and grid assembly.
Subject: Easiest tube amplifier possible
I read and enjoyed your article on the simplest possible amplifier. I was thinking of building it using a pair of One Electron output transformers. My question is: Wouldn't a five 6BQ5 stage need an se output transformer that could handle something like 200 mA of current without saturating? I'm absolutely a novice at this so bear with me?
Bob
It all depends on how much idle current each 6BQ5 will be expected to conduct. Triode connected, the 6BQ5 has a maximum dissipation of watts. Given a power supply voltage of 360 volts, roughly the result of rectifying a 240 VAC transformer winding, the maximum permissible idle current draw per 6BQ5 would be 38 mA (14 watts / 360 volts). So, 32 mA would be safe, particularly as the cathode biasing and the voltage drop through the output transformer's primary winding would subtract from the total cathode-to-plate voltage. Five 6BQ5s would then draw a total of 160 mA of current, which is the UBT-1's limit. The load each 6BQ5 would see would equal the transformer's 1600 primary impedance times five, or 8,000 ohms. All in all, I think this would work well.
|
|
|
|
|
